## 红楼梦文本挖掘之数据预处理####
## 主要用于文本文档的读取和构建
## 分析与挖掘R中的人物关系
## 红楼梦中关键人物的社交网络
## 孙玉林；2016年10月31
## 如果在每个段落中人物同时出现，则频数权重加1
## 加载所需要的包
library(igraph)
##
## Attaching package: 'igraph'
## The following objects are masked from 'package:stats':
##
## decompose, spectrum
## The following object is masked from 'package:base':
##
## union
library(ggplot2)
library(gridExtra)
library(networkD3)
## 读取数据
## 一共有149人出现的频次多余10次
Red_net <- read.csv("./数据/社交网络权重.csv")
Red_net[,1:2] <- apply(Red_net[,1:2],2,as.character)
Name_freq <- read.csv("./数据/红楼梦人物出现频次.csv")
Name_freq <- Name_freq[Name_freq$word %in%(union(unique(Red_net$First),
unique(Red_net$Second))),]
Name_freq$word <- as.character(Name_freq$word)
# union(unique(Red_net$First),unique(Red_net$Second))
## 可视化人出现的频次
p1 <- ggplot(Name_freq,aes(x = reorder(word,freq),y = freq)) +
theme_bw(base_size = 9,base_family = "STKaiti")+
geom_bar(stat = "identity",position = "dodge",fill = "lightblue") +
theme(axis.text.x = element_text(size = 5,hjust = 1,angle = 90,vjust = 0.5),
axis.title.x = element_blank()) +
labs(x = "",y = "频数",title = "《红楼梦》中关键人物出现次数")
p2 <- ggplot(Name_freq[Name_freq$freq>80,],aes(x = reorder(word,freq),y = freq)) +
theme_bw(base_size = 9,base_family = "STKaiti")+
geom_bar(stat = "identity",position = "dodge",fill = "lightblue") +
theme(axis.text.x = element_text(size = 9,hjust = 1,angle = 90,vjust = 0.5)) +
labs(x = "人名",y = "频数")
grid.arrange(p1,p2,nrow = 2)
## 按照权重1，即章节权重分析人物的社交网络####
# 1:准备社交网络数据
chap_net <- Red_net[Red_net$chapweight > 10,c(1,2,3)]
names(chap_net) <- c("from","to","weight")
chap_vert <- Name_freq[Name_freq$word %in% as.character(union(unique(chap_net$from),
unique(chap_net$to))),]
chap_net <- graph_from_data_frame(chap_net,directed = FALSE,
vertices = chap_vert)
chap_net
## IGRAPH UNW- 69 762 --
## + attr: name (v/c), freq (v/n), weight (e/n)
## + edges (vertex names):
## [1] 宝玉--贾母 宝玉--凤姐 宝玉--袭人 宝玉--王夫人
## [5] 宝玉--宝钗 宝玉--贾政 宝玉--贾琏 宝玉--平儿
## [9] 宝玉--薛姨妈 宝玉--探春 宝玉--紫鹃 宝玉--鸳鸯
## [13] 宝玉--贾珍 宝玉--李纨 宝玉--尤氏 宝玉--晴雯
## [17] 宝玉--邢夫人 宝玉--薛蟠 宝玉--林黛玉 宝玉--香菱
## [21] 宝玉--麝月 宝玉--贾蓉 宝玉--贾赦 宝玉--惜春
## [25] 宝玉--贾芸 宝玉--周瑞家的 宝玉--芳官 宝玉--贾环
## [29] 宝玉--妙玉 宝玉--雪雁 宝玉--迎春 宝玉--赵姨娘
## + ... omitted several edges
chap_net$name <- "《红楼梦》章节人物关系"
V(chap_net)$media
## NULL
## 节点数目
vcount(chap_net)
## [1] 69
## 边的数目
ecount(chap_net)
## [1] 762
## 简化网络图
chap_net <- simplify(chap_net,remove.multiple = TRUE,remove.loops = TRUE,
edge.attr.comb = "mean")
## 查看节点的度
degrees <- data.frame(name = names(degree(chap_net)),
counts = (degree(chap_net)))
ggplot(degrees,aes(x = reorder(name,counts),y = counts)) +
theme_bw(base_size = 11,base_family = "STKaiti")+
geom_bar(stat = "identity",position = "dodge",fill = "lightblue") +
theme(axis.text.x = element_text(size = 5,hjust = 1,angle = 90,vjust = 0.5),
axis.title.x = element_blank()) +
labs(x = "人名",y = "节点的度",title = "《红楼梦》")
## 判断事否为联通图
is.connected(chap_net)
## [1] TRUE
## 计算图的直径
diameter(chap_net,directed = FALSE)
## [1] 48
##
set.seed(1234)
par(cex = 0.8,family = "STKaiti")
## 设置图层
layout1 <- layout.lgl(chap_net)
layout2 <- layout.kamada.kawai(chap_net)
layout3 <- layout.reingold.tilford(chap_net)
layout4 <- layout.fruchterman.reingold(chap_net)
#V(chap_net)$size <- Name_freq$freq/10
## 设置节点的字体
V(chap_net)$label.family <- "STKaiti"
E(chap_net)$width <- log10(E(chap_net)$weight) *2
egam <- (E(chap_net)$width) / max(E(chap_net)$width)
E(chap_net)$color <- rgb(1,0.5,0.5,egam)
V(chap_net)$size <- log(V(chap_net)$freq) * 2.5
plot(chap_net,layout = layout1,main = "《红楼梦》根据章节部分人物关系")
plot(chap_net,layout = layout2,main = "《红楼梦》根据章节部分人物关系")
plot(chap_net,layout = layout3,main = "《红楼梦》根据章节部分人物关系")
plot(chap_net,layout = layout4,main = "《红楼梦》根据章节部分人物关系")
## 按照权重2，即段落权重分析人物的社交网络####
# 1:准备社交网络数据
chap_net <- Red_net[Red_net$duanweight > 10,c(1,2,4)]
names(chap_net) <- c("from","to","weight")
chap_vert <- Name_freq[Name_freq$word %in% as.character(union(unique(chap_net$from),
unique(chap_net$to))),]
chap_net <- graph_from_data_frame(chap_net,directed = FALSE,
vertices = chap_vert)
chap_net
## IGRAPH UNW- 77 398 --
## + attr: name (v/c), freq (v/n), weight (e/n)
## + edges (vertex names):
## [1] 宝玉--贾母 宝玉--凤姐 宝玉--袭人 宝玉--王夫人 宝玉--宝钗
## [6] 宝玉--贾政 宝玉--贾琏 宝玉--平儿 宝玉--薛姨妈 宝玉--探春
## [11] 宝玉--紫鹃 宝玉--鸳鸯 宝玉--贾珍 宝玉--李纨 宝玉--尤氏
## [16] 宝玉--晴雯 宝玉--刘姥姥 宝玉--邢夫人 宝玉--薛蟠 宝玉--林黛玉
## [21] 宝玉--香菱 宝玉--麝月 宝玉--贾蓉 宝玉--贾赦 宝玉--惜春
## [26] 宝玉--贾芸 宝玉--芳官 宝玉--贾环 宝玉--妙玉 宝玉--雪雁
## [31] 宝玉--迎春 宝玉--赵姨娘 宝玉--莺儿 宝玉--秦钟 宝玉--巧姐
## [36] 宝玉--秋纹 宝玉--贾兰 宝玉--茗烟 宝玉--史湘云 宝玉--大了
## + ... omitted several edges
chap_net$name <- "《红楼梦》章节人物关系"
V(chap_net)$media
## NULL
## 节点数目
vcount(chap_net)
## [1] 77
## 边的数目
ecount(chap_net)
## [1] 398
## 简化网络图
chap_net <- simplify(chap_net,remove.multiple = TRUE,remove.loops = TRUE,
edge.attr.comb = "mean")
## 查看节点的度
degrees <- data.frame(name = names(degree(chap_net)),
counts = (degree(chap_net)))
ggplot(degrees,aes(x = reorder(name,counts),y = counts)) +
theme_bw(base_size = 11,base_family = "STKaiti")+
geom_bar(stat = "identity",position = "dodge",fill = "lightblue") +
theme(axis.text.x = element_text(size = 5,hjust = 1,angle = 90,vjust = 0.5),
axis.title.x = element_blank()) +
labs(x = "人名",y = "节点的度",title = "《红楼梦》")
## 判断事否为联通图
is.connected(chap_net)
## [1] TRUE
## 计算图的直径
diameter(chap_net,directed = FALSE)
## [1] 87
##
set.seed(1234)
par(cex = 0.8,family = "STKaiti")
## 设置图层
layout1 <- layout.lgl(chap_net)
layout2 <- layout.kamada.kawai(chap_net)
layout3 <- layout.reingold.tilford(chap_net)
layout4 <- layout.fruchterman.reingold(chap_net)
#V(chap_net)$size <- Name_freq$freq/10
## 设置节点的字体
V(chap_net)$label.family <- "STKaiti"
E(chap_net)$width <- log10(E(chap_net)$weight) *2
egam <- (E(chap_net)$width) / max(E(chap_net)$width)
E(chap_net)$color <- rgb(1,0.5,0.5,egam)
V(chap_net)$size <- log(V(chap_net)$freq) * 2.5
plot(chap_net,layout = layout1,main = "《红楼梦》根据段落部分人物关系")
plot(chap_net,layout = layout2,main = "《红楼梦》根据段落部分人物关系")
plot(chap_net,layout = layout3,main = "《红楼梦》根据段落部分人物关系")
plot(chap_net,layout = layout4,main = "《红楼梦》根据段落部分人物关系")
## 按照权重2，即段落权重分析人物的社交网络####
## 分析链接次数较大的人物
# 1:准备社交网络数据
chap_net <- Red_net[Red_net$duanweight > 50,c(1,2,4)]
names(chap_net) <- c("from","to","weight")
chap_vert <- Name_freq[Name_freq$word %in% as.character(union(unique(chap_net$from),
unique(chap_net$to))),]
chap_net <- graph_from_data_frame(chap_net,directed = FALSE,
vertices = chap_vert)
chap_net
## IGRAPH UNW- 24 70 --
## + attr: name (v/c), freq (v/n), weight (e/n)
## + edges (vertex names):
## [1] 宝玉 --贾母 宝玉 --凤姐 宝玉 --袭人 宝玉 --王夫人
## [5] 宝玉 --宝钗 宝玉 --贾政 宝玉 --贾琏 宝玉 --平儿
## [9] 宝玉 --薛姨妈 宝玉 --探春 宝玉 --紫鹃 宝玉 --鸳鸯
## [13] 宝玉 --贾珍 宝玉 --李纨 宝玉 --尤氏 宝玉 --晴雯
## [17] 宝玉 --邢夫人 宝玉 --林黛玉 宝玉 --麝月 宝玉 --惜春
## [21] 宝玉 --迎春 贾母 --凤姐 贾母 --袭人 贾母 --王夫人
## [25] 贾母 --宝钗 贾母 --贾政 贾母 --贾琏 贾母 --薛姨妈
## [29] 贾母 --探春 贾母 --鸳鸯 贾母 --贾珍 贾母 --李纨
## + ... omitted several edges
chap_net$name <- "《红楼梦》章节人物关系"
V(chap_net)$media
## NULL
## 节点数目
vcount(chap_net)
## [1] 24
## 边的数目
ecount(chap_net)
## [1] 70
## 简化网络图
chap_net <- simplify(chap_net,remove.multiple = TRUE,remove.loops = TRUE,
edge.attr.comb = "mean")
## 查看节点的度
degrees <- data.frame(name = names(degree(chap_net)),
counts = (degree(chap_net)))
ggplot(degrees,aes(x = reorder(name,counts),y = counts)) +
theme_bw(base_size = 11,base_family = "STKaiti")+
geom_bar(stat = "identity",position = "dodge",fill = "lightblue") +
theme(axis.text.x = element_text(size = 5,hjust = 1,angle = 90,vjust = 0.5),
axis.title.x = element_blank()) +
labs(x = "人名",y = "节点的度",title = "《红楼梦》")
## 判断事否为联通图
is.connected(chap_net)
## [1] TRUE
## 计算图的直径
diameter(chap_net,directed = FALSE)
## [1] 289
##
set.seed(1234)
par(cex = 0.8,family = "STKaiti")
## 设置图层
layout1 <- layout.lgl(chap_net)
layout2 <- layout.kamada.kawai(chap_net)
layout3 <- layout.reingold.tilford(chap_net)
layout4 <- layout.fruchterman.reingold(chap_net)
#V(chap_net)$size <- Name_freq$freq/10
## 设置节点的字体
V(chap_net)$label.family <- "STKaiti"
E(chap_net)$width <- log10(E(chap_net)$weight) *2
egam <- (E(chap_net)$width) / max(E(chap_net)$width)
E(chap_net)$color <- rgb(1,0.5,0.5,egam)
V(chap_net)$size <- log(V(chap_net)$freq) * 2.5
plot(chap_net,layout = layout1,main = "《红楼梦》根据段落部分人物关系")
plot(chap_net,layout = layout2,main = "《红楼梦》根据段落部分人物关系")
plot(chap_net,layout = layout3,main = "《红楼梦》根据段落部分人物关系")
plot(chap_net,layout = layout4,main = "《红楼梦》根据段落部分人物关系")
library(networkD3)
library(igraph)
# Basic Graph
chap_net <- Red_net[Red_net$duanweight > 40,c(1,2,4)]
g <- graph.data.frame(chap_net, directed=F) # raw graph
## Make a vertices df
vertices<-data.frame(
name = V(g)$name,
group = edge.betweenness.community(g)$membership,
betweenness = (betweenness(g,directed=F,normalized=T)*115)+0.1 #so size isn't tiny
)
#nb. can also adjust nodesize with `radiusCalculation`
# create indices (indexing needs to be JS format)
chap_net$source.index = match(chap_net$First, vertices$name)-1
chap_net$target.index = match(chap_net$Second, vertices$name)-1
# supply a edgelist + nodelist
d3 = forceNetwork(Links = chap_net, Nodes = vertices,
Source = 'source.index', Target = 'target.index',
NodeID = 'name',
Group = 'group', # color nodes by group calculated earlier
charge = -200, # node repulsion
linkDistance = 20,
zoom = T,
opacity = 1,
fontSize=24)
show(d3)
# Basic Graph
chap_net <- Red_net[Red_net$duanweight > 60,c(1,2,4)]
g <- graph.data.frame(chap_net, directed=F) # raw graph
## Make a vertices df
vertices<-data.frame(
name = V(g)$name,
group = edge.betweenness.community(g)$membership,
betweenness = (betweenness(g,directed=F,normalized=T)*115)+0.1 #so size isn't tiny
)
#nb. can also adjust nodesize with `radiusCalculation`
# create indices (indexing needs to be JS format)
chap_net$source.index = match(chap_net$First, vertices$name)-1
chap_net$target.index = match(chap_net$Second, vertices$name)-1
# supply a edgelist + nodelist
d3 = forceNetwork(Links = chap_net, Nodes = vertices,
Source = 'source.index', Target = 'target.index',
NodeID = 'name',
Group = 'group', # color nodes by group calculated earlier
charge = -200, # node repulsion
linkDistance = 20,
zoom = T,
opacity = 1,
fontSize=24)
show(d3)